Position Effect and AUelism in Drosophila 



191 



have the best chance of yielding a positive re- 

 sult if the two pairs of genes concerned are 

 adjacent or very close to each other. But 

 this also means that if the genes are very close 

 together crossing over will rarely occur be- 

 tween them, and large numbers of progeny 

 will have to be observed to be sure at least 

 one exchange between them has been in- 

 cluded among the results! What particular 

 genes should we use in our test? Should we 

 use linear neighbors that apparently perform 

 very different functions, or shall we use those 

 having similar effects? If the effect of one 

 gene is to be modified by the effect of its 

 linear neighbor, it would seem reasonable to 

 use neighboring genes that have similar ef- 

 fects, for in this case changing the allele of the 

 neighboring gene might be expected to pro- 

 duce a change in its own effect. Suppose, 

 on the other hand, the two neighbor genes 

 affected totally different traits. In this case, 

 while changing the allele of the neighboring 

 gene would mean that the neighboring allele 

 now present affects the totally different trait 

 in a different way, this would not be expected 

 to have any effect on the functioning of its 

 linear neighbor. 



The best source of genes which have very 

 similar effects is, of course, the members of a 

 multiple allelic series, for instance, the mem- 

 bers of the white series on the X chromo- 

 some of Drosophila. But you immediately 

 might say that these cannot be used in our ex- 

 periment, since only one is present on each 

 homolog, and we need two pairs of genes to 

 perform a cis-trans test for position effect. 

 Moreover, all these alleles of white are located 

 at 1.5 on the crossover map. When the 

 "trans" hybrid for w {apricot) and w {white) 

 is made, it is found to produce pale apricot eye 

 color. However, such a phenotype does not 

 prove w" and w are alleles. Alleles are alter- 

 natives of the same gene; genes are most 

 properly identified as alleles because they oc- 

 cupy the same site on homologous chromo- 

 somes and cannot be recombined as the con- 



sequence of crossing over, and because, bar- 

 ring nondisjunction, the members of a pair of 

 alleles always segregate. The statement that 

 w° and w are located at position 1.5 may be 

 incorrect! Suppose these two are really non- 

 allelic but similarly acting genes located close 

 together, one at position 1.49 {w") and one at 

 1.51 (w). The crossover data, being finite 

 and somewhat variable, could have acciden- 

 tally placed them both at locus 1.5. If vv 

 and w are, in fact, close but not allelic to each 

 other, the trans heterozygote for them should 

 yield the cis heterozygote by crossing over. 

 But if these genes are only .02 of a crossover 

 unit apart, only one such crossover would oc- 

 cur among 5,000 tested chromosomes. If only 

 a few hundred flies are scored in a search for 

 such crossovers, it is very likely none of this 

 type will be found. Such an experiment 

 would have to be done on a very large scale 

 to serve as a test for nonallelism. 



Let us examine the plan and results of a 

 large scale experiment ^ actually designed to 

 test the nonallelism of h'" and w. Drosophila 

 females were constructed carrying an at- 

 tached-X chromosome containing y w spl on 

 one arm and j+ h-" and 57;/+ on the other. Re- 

 call that the use of attached-X's permits one 

 to recover two strands of the four involved 

 in each crossover (cf. p. 122). The present 

 genetic system even permits both comple- 

 mentary crossover types to appear simul- 

 taneously in the same gamete. The left part 

 of Figure 22-5 shows schematically a portion 

 of this attached-X as it would appear in the 

 tetrad stage at the time of the crossing over, 

 and indicates the standard genetic map loca- 

 tion of the y and spl markers. The female 

 carrying this chromosome has a pale 

 (dilute) apricot phenotype. If such a female 

 is crossed to a 5a/--containing male, the non- 

 Bar Fi daughters (who carry a Y from their 

 father) are usually noncrossovers and have 

 pale apricot-colored eyes like their mother. 

 Crossovers which occur between the white 

 ^ Based upon work of E. B. Lewis. 



